Patients with cirrhosis whose first episode of acute variceal bleeding occurs when they are already taking prophylactic beta-blockers are at increased risk for recurrent bleeding, Dr. Andrea Ribeiro de Souza and colleagues reported in the June issue of Clinical Gastroenterology and Hepatology.
The risk of recurrence is approximately twice as high in such patients as in those who are not taking prophylactic beta-blockers when their first variceal bleed occurs. This is true even when patients receive the currently recommended secondary therapy after nonselective beta-blocker prophylaxis fails, which is a combination of endoscopic band ligation and further beta-blocker treatment, with or without the addition of isosorbide-5-mononitrate.
These results, taken together with those of two recent studies showing that patients who undergo endoscopic band ligation have a "dismal" rate of variceal rebleeding, suggest that patients who don’t respond to prophylactic beta-blockers "have an idiosyncrasy that makes them also poor responders to endoscopic therapy.
"Since there are no baseline clinical or hemodynamic characteristics that could differentiate this population, it can be speculated that their increased bleeding risk may be related to other factors, perhaps ... peculiarities of the esophageal circulation, which [have] never been investigated so far," said Dr. de Souza and associates at the University of Barcelona and Centro de Investigacion Biomedica en Red de Enfermedades Hepaticas y Digestivas (Ciberehd) (Clin. Gastroenterol. Hepatol. 2012 [doi:10.1016/j.cgh.2012.02.011]).
Primary prophylaxis of variceal bleeding with nonselective beta-blockers is now widely used, so the number of cirrhosis patients who experience their first episode of bleeding while taking these drugs is increasing. Until now, no study has explored whether these patients differ significantly from those who aren’t taking the drugs when they have their first variceal bleed.
Dr. de Souza and colleagues examined this question using data from the liver unit of their hospital during 2007-2011, on 89 consecutive patients treated for acute variceal bleeding. Thirty-four of the study subjects had their first bleed while on beta-blocker prophylaxis, and 55 subjects were not taking the medication.
Subjects were treated according to current recommendations. On admission they received an intravenous vasoconstrictor (terlipressin or somatostatin) and prophylactic antibiotics, and they underwent endoscopic band ligation (EBL) within 12 hours. Those whose bleeding was controlled were started on oral propanolol or nadolol, which was increased until heart rate or systolic blood pressure had fallen to appropriate levels.
Isosorbide was started in 21 patients. EBL sessions were scheduled every 2 weeks until varices were eradicated, and patients took proton pump inhibitors until that time as well.
Variceal obliteration was achieved in only 67% of patients who had already been taking beta-blocker prophylaxis, compared with 80% of those who had not.
All subjects underwent surveillance endoscopy at 1-3 months, and at 6-month intervals thereafter. Further EBL was done if varices reappeared. Patients were followed for 2 years, or until liver transplantation or death occurred.
The primary end point of the study was rebleeding from any source during follow-up. The cumulative incidence of rebleeding from any source was 48% for patients already taking beta-blocker prophylaxis, compared with 24% in the other group.
When the analysis was restricted to rebleeding from varices only, the rate was still significantly higher among patients already taking beta-blocker prophylaxis (39%) than in the other group (17%).
This discrepancy persisted across all subgroups in further analyses, regardless of whether the cirrhosis was or was not alcohol related, whether or not the subjects were actively drinking at the time of the first variceal bleed, and whether or not patients were treated with isosorbide.
These findings indicate that patients whose first variceal bleed occurs while they are taking prophylactic beta-blockers are not likely to benefit from EBL, "and would probably be best treated with new and more effective drugs to achieve target reductions in portal pressure. A possibility is the use of carvedilol, a nonselective beta-blocker with intrinsic vasodilator activity that causes a greater reduction in hepatic vein pressure gradient than propranolol or nadolol," the researchers said.
An even better option might be transjugular intrahepatic portosystemic shunting, since medication typically achieves only a modest decrease in hepatic vein pressure gradient, which may not be sufficient to prevent bleeding recurrences, they added.
The study was supported in part by grants from Instituto de Salud Carlos III, Ministerio de Ciencia e Innovación. The Ciberehd is funded by Instituto de Salud Carlos III. Dr. Andrea Ribeiro de Souza’s work is funded by grant of the BBVA foundation. The investigators reported no financial conflicts of interest.